A Texture based Tumor detection and automatic Segmentation using Seeded Region Growing Method
Abstract
Detection and segmentation of Brain tumor accurately is a challenging task in MRI. The MRI image is an image that produces a high contrast images indicating regular and irregular tissues that help to distinguish the overlapping in margin of each limb. All automatic seed finding methods may suffer with the problem if there is no growth of tumor and any small white part is there. But when the edges of tumor is not sharped then the segmentation results are not accurate i.e. segmentation may be over or under. This may be happened due to initial stage of the tumors [5]. So , in this paper a method of tumor detection based on texture of the MRI and if it is detected then to segment it automatically is proposed in this paper to separate the irregular from the regular surrounding tissue to get a real identification of involved and noninvolved area that help the surgeon to distinguish the involved area precisely. The method used in this paper is texture analysis and seeded region growing method and it was implemented using MATLAB 7.6.0.324 on 25 Magnetic Resonance Images having brain tumors and also on images without any abnormality to detect the tumor boundaries in 2D MRI for different cases.
Index Terms— Gray level, MRI image, Region growing, tumour, segmentation, Texture analysis etc.
1. Introduction
Segmentation is a process of identifying an object or pattern in the given work space. In this project we are considering magnetic resonance image as our work space. Actually the MRI produces a high contrast image representing each part very clearly, but sometimes due to be determined accurately so a problem of segmenting it is always there. In these cases the physiologist always need to have keen observation of the anatomical structure. But this process is too much time consuming and if the initial segmentation result is not correct then other consequent results
like volume calculation also produces incorrect measurement results.
There are a number of methods for brain tumor segmentation like fuzzy logic approach, neuro fuzzy approach, Random walk etc, but these all methods can produces unsatisfactory results due to unsharped edge boundaries and also the time to produce desire result is large[6] . In this paper we are proposing a texture based analysis to detect abnormality in the brain and an automatic region growing method to segment the brain tumours. In this paper we are combining the two parameters to produce more accurate results. Also in this method the users don’t need to select the seed point manually therefore there is no need of human intervention [2]. In this project work our assumption is that the brain tumor have grown in considerable size and their structure may be of any type like snakelike or circular shaped etc[1].
2. Proposed method
In this paper our method proposed has divided into four subparts. The output obtained from one part is taken as input to the next part. This can be represented by following work flow graph:


Download full report
http://googleurl?sa=t&source=web&cd=1&ve...020422.pdf&ei=sMZVTtbSNseyrAfC5NieCw&usg=AFQjCNH9BJgD1CnJBST6WgE-aWuFIi26cQ&sig2=41CnZAFsZCHylGeaFqdZeQ